DEXA vs BIA Accuracy
Why your £100 smart scale disagrees with a £100,000 clinical scanner. We break down the variance and margins of error between DEXA and Bioelectrical Impedance.
The Illusion of Precision
Most people own a bioelectrical impedance (BIA) scale, stepping on it daily to log a body fat percentage down to a single decimal point (e.g., 14.3%). But precision is not accuracy. That 14.3% might actually represent a true physical state of 11%, or 18%.
When athletes and data-driven individuals graduate to clinical Dual-Energy X-Ray Absorptiometry (DEXA) scans, they are often shocked. The smart scale might say 12%, but the DEXA scan reads 17%. Why?
Bioelectrical Impedance (BIA)
BIA works by sending a small electrical current through the body.
Muscle tissue contains roughly 70-75% water and is highly conductive. Adipose tissue (fat) contains very little water and heavily resists the electrical current. The scale measures the impedance (resistance), factors in your height and weight, and runs it through a proprietary algorithm to estimate fat-free mass.
Common Errors in BIA
- Hydration Status: Dehydration increases resistance. If you are dehydrated, the scale thinks you have less muscle (water) and therefore more fat.
- Carbohydrate Depletion: Glycogen stores water in muscle (3g of water per 1g of glycogen). A low-carb diet depletes glycogen, causing the scale to read a massive loss of muscle mass and an artificial spike in body fat percentage.
- Foot-to-Foot Limitations: Standard smart scales only send current up one leg and down the other. They never measure your torso or arms, entirely guessing upper body composition based on population averages.
- Algorithm Bias: BIA algorithms often use your age and gender inputs to heavily skew the result toward the population mean. If you are an outlier (e.g., heavily muscled), the algorithm will misinterpret your high body mass index.
Error Margin: Studies show consumer BIA scales have an error margin of ±4% to ±8% when compared to a 4-compartment model.
Dual-Energy X-Ray Absorptiometry (DEXA)
Originally designed to measure bone mineral density to diagnose osteoporosis, DEXA is a 3-compartment clinical model. It measures:
- Bone Mineral Content (BMC)
- Fat Mass (FM)
- Lean Tissue Mass (LTM)
DEXA uses two different energy levels of X-rays. Tissue absorbs these X-rays differently based on density. By measuring the attenuation (loss of energy) of the X-rays as they pass through the body, the machine builds a pixel-by-pixel map of your tissue.
Why DEXA Reads "Higher"
DEXA almost always reports a higher body fat percentage than calipers or BIA. This is primarily because DEXA is highly accurate at measuring visceral fat (fat wrapped around organs) and intramuscular fat (fat marbled within muscle tissue). Calipers only measure subcutaneous fat (under the skin), and BIA algorithms often underpredict visceral loads.
Error Margin: Even DEXA is not perfect. Its error margin is roughly ±1.5% to ±2.5%. Variations occur based on the machine's manufacturer (GE Lunar vs Hologic), the software version, and the subject's hydration status, as DEXA assumes a constant hydration fraction of lean mass (73.2%).
Comparison Table
| Metric | BIA (Smart Scales) | DEXA Scan |
|---|---|---|
| Underlying Mechanism | Electrical resistance | X-ray attenuation |
| Model Type | 2-Compartment (guessing) | 3-Compartment |
| Measures Visceral Fat? | No (Algorithmic guess) | Yes |
| Regional Data (Arms/Legs)? | Only on 8-point models | Yes (Highly precise) |
| Typical Error Margin | ±5.0% | ±2.0% |
| Cost Per Test | £0 (after hardware purchase) | £100 - £150 |
The Protocol for BIA Users
If you refuse to pay for DEXA and want to track trends using a BIA scale, you must standardise your environment to minimize the massive error bars:
- Test Fasted: Wake up, use the toilet, and weigh yourself immediately before eating or drinking.
- Stable Glycogen: Do not compare a reading from a high-carb week to a reading from a low-carb week.
- Ignore the Absolute Number: Focus only on the trend. If the BIA scale goes from 18% to 16% over 12 weeks while your strength goes up, you lost fat, regardless of whether the true number is 14% or 20%.